Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    legal/emergency situations

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    Tudo
    -0.08
     everything
    -0.07
     hadir
    -0.07
    .Double
    -0.07
    Tout
    -0.07
    .Null
    -0.07
     Everything
    -0.07
    Toutes
    -0.07
    祝
    -0.07
    ও
    -0.07
    POSITIVE LOGITS
     adversely
    0.13
     harmed
    0.11
     jeopard
    0.11
     bezpieczeń
    0.10
     harm
    0.10
     harms
    0.09
     sécurité
    0.09
     harming
    0.09
     incapac
    0.09
     prevention
    0.09
    Act Density 0.052%

    No Known Activations